Friday, 20 January 2017

The tip of the iceberg?

It has been suggested that some people with Multiple Sclerosis have subtle evidence of disease activity long before they develop clinically-definite disease. The argument goes that if you performed detailed examinations and investigations, you would detect these subtle abnormalities in certain people before they meet diagnostic criteria for MS. In fact, according to one estimate, about 1/3 of people who have evidence of MS on neuroimaging go on to develop MS within 5 years.

We also now have good evidence that early, aggressive treatment improves long-term outcomes for people with Clinically Isolated Syndrome (‘pre-MS’). It is theoretically plausible that treating people even earlier – e.g. when they have abnormalities on brain imaging but no clinical signs or symptoms of MS – could reduce their risk of developing MS proper.

If preventing MS is going to be an achievable goal for the future, it is crucial that we are able to determine who is at risk of developing the disease so that we can target our therapy appropriately. The Genes and Environment in Multiple Sclerosis (GEMS) project has set out to do just this – it recruited 2632 1st degree relatives of people with MS to get a better understanding of how to predict the development and the progression of this protean disease. 

In a new paper, the GEMS investigators asked whether their MS risk score could predict who would have subtle clinical or radiological abnormalities suggestive of MS. To do this, they took a subset of 100 people from their study cohort, calculated each person’s risk using their previously published risk score, and then took the people in the top and bottom 10% of risk scores to investigate in more detail. They hypothesised that there would be a difference in clinical and radiological parameters relevant to MS between the ‘high’ and ‘low’ risk groups.

To avoid a gender bias, they focussed their analysis on women only. They compared 40 ‘high risk’ women to 25 ‘low risk’ women. Despite using a battery of clinical and imaging measures, they found no statistically significant differences between the groups in terms of age, height, cigarette smoking, vitamin D level, history of infectious mononucleosis, history of migraine, number of MRI lesions suggestive of MS, the prevalence of radiologically-isolated syndrome, EDSS, NHPT, or PASAT scores. There was, however, a significant difference in vibration sensitivity between the high and low risk groups, with the high risk group demonstrating a diminished sensitivity to vibration.

This study demonstrates nicely how difficult it is to detect subtle abnormalities in a small cohort. Even when comparing the highest and lowest risk groups as stratified by the GEMS risk score, only one parameter – vibration sensitivity – was significantly different between the groups. This is unsurprising. If early, ‘pre-MS’ is defined by subtle clinical or imaging abnormalities, then a big cohort is required to demonstrate this. When trying to study small effects, we need big numbers to reduce the chances of false negative results. The cohort of 60 or so people used in this study is simply not big enough to meaningfully prove or disprove the hypothesis at hand. I would take their lack of positive findings with a pinch of salt. 

The motivation behind the GEMS project is really laudable, and I imagine that lots of interesting data will come out as they follow their cohort over the next few years. Armed with a better understanding of who is at risk of MS, the paradigm may begin to shift from early treatment to prevention. Part of the problem is that it is notoriously difficult to demonstrate benefit in prevention trials as the number of people who will develop the disease without any treatment is small. Once we work out who is at high risk of developing MS we will be much better placed to design clinical trials that could show meaningful benefit for preventative therapies. 

The Paper

Importance  Subclinical inflammatory demyelination and neurodegeneration often precede symptom onset in multiple sclerosis (MS).
Objective  To investigate the prevalence of brain magnetic resonance imaging (MRI) and subclinical abnormalities among asymptomatic individuals at risk for MS.
Design, Setting, and Participants  The Genes and Environment in Multiple Sclerosis (GEMS) project is a prospective cohort study of first-degree relatives of people with MS. Each participant’s risk for MS was assessed using a weighted score (Genetic and Environmental Risk Score for Multiple Sclerosis Susceptibility [GERSMS]) comprising an individual’s genetic burden and environmental exposures. The study dates were August 2012 to July 2015.
Main Outcomes and Measures  Participants in the top and bottom 10% of the risk distribution underwent standard and quantitative neurological examination, including disability status, visual, cognitive, motor, and sensory testing, as well as qualitative and quantitative neuroimaging with 3-T brain MRI and optical coherence tomography.
Results  This study included 100 participants at risk for MS, with 41 at higher risk (40 women [98%]) and 59 at lower risk (25 women [42%]), at a mean (SD) age of 35.1 (8.7) years. Given the unequal sex distribution between the 2 groups, the analyses were restricted to women (n = 65). When considering all measured outcomes, higher-risk women differed from lower-risk women (P = .01 by omnibus test). Detailed testing with a vibration sensitivity testing device in a subgroup of 47 women showed that higher-risk women exhibited significantly poorer vibration perception in the distal lower extremities (P = .008, adjusting for age, height, and testing date). Furthermore, 5 of 65 women (8%) (4 at higher risk and 1 at lower risk) met the primary neuroimaging outcome of having T2-weighted hyperintense brain lesions consistent with the 2010 McDonald MRI criteria for dissemination in space. A subset of participants harbor many different neuroimaging features associated with MS, including perivenous T2-weighted hyperintense lesions and focal leptomeningeal enhancement, consistent with the hypothesis that these individuals are at higher risk of developing clinical symptoms of MS than the general population.
Conclusions and Relevance  Higher-risk asymptomatic family members of patients with MS are more likely to have early subclinical manifestations of MS. These findings underscore the importance of early detection in high-risk individuals.
Trial Registration Identifier: NCT01353547

1 comment:

  1. Dear Dr Ben,

    This is disappointing, I was hoping that this would be a data problem. I was hoping that we could combine readings from n tests and see some kind of pattern in the data. It's hard to believe that there is no signal of a problem in advance of a clinical presentation.


Please note that all comments are moderated and any personal or marketing-related submissions will not be shown.